Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Julius Han · Gated SAE for Llama3-8B-Instruct
    3. Llama3-8B-IT
    4. Residual Stream
    5. 25-RES-JH
    6. 29429
    Prev
    Next
    INDEX
    Explanations

    references to scientific measurements related to psychotic conditions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    Juliushanhanhan/llama-3-8b-it-res/blocks.25.hook_resid_post
    Features
    65,536
    Data Type
    float32
    Hook Name
    blocks.25.hook_resid_post
    Hook Layer
    25
    Architecture
    gated
    Context Size
    1,024
    Dataset
    Juliushanhanhan/openwebtext-1b-llama3-tokenized-cxt-1024
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ะ
    -0.16
    .ExecuteReader
    -0.15
    etag
    -0.15
    regnum
    -0.14
     GLenum
    -0.14
    _argv
    -0.14
    Č↵
    -0.13
    óż
    -0.13
    _errno
    -0.13
     souvenir
    -0.13
    POSITIVE LOGITS
     EH
    0.20
    (EC
    0.20
    (E
    0.20
     EE
    0.18
    _ES
    0.18
     EV
    0.18
    ET
    0.18
     EW
    0.18
    [E
    0.18
     ES
    0.17
    Activations Density 0.053%

    No Known Activations